Cylinder Configuration Research Articles

Dual-functional synergetic energy harvesting and flow-induced vibration control of an electromagnetic-based square cylinder integrated with a flexible bimorph piezoelectric wake splitter plate

A dual-purpose FIV-based hydroelastic energy harvesting and cylinder response suppression strategy that functions based on the synergy of piezoelectric and electromagnetic transduction (EMT) mechanisms is proposed and numerically implemented. The hybrid harvester consists of a linearly sprung (1DOF) square cylinder fitted on the wake side with a thin flexural-mode cantilever bimorph piezoelectric (PVDF) splitter plate in real-time collaboration with a transversely hooked induction-based magnet-coil type transducer. The Reynolds averaged Navier–Stokes (RANS) equations with the shear stress transport (SST) k-ω turbulence closure model are selected for qualitative/quantitative prediction of hydrodynamic flow behavior in a relatively wide Reynolds numbers range. Numerical simulations show that increasing Reynolds number for the single-alone EMT-equipped cylinder in the low to intermediate range (2×103≤Re≤3×104) can noticeably improve the system hydrokinetic energy harvesting performance where a distinct coupled VIV/galloping effect is observed. Also, the hybrid piezoelectromagnetic harvester is capable of effectively suppressing the key response parameters and considerably increase the total system electrical output in an extended working bandwidth (3×104<Re≤4×105). In particular, in contrast to the conventional fixed-base cylinder configurations, maximum power extraction can be achieved over most span of the flexible piezoelectric splitter plate at Re=4×105 where the flapping oscillations of splitter plate synchronizes with the downstream alternating wake vortices.

Numerical investigation of the role of linear and nonlinear forces in determining the direction of electric wind caused by atmospheric pressure DC corona discharge

In this study, the force generated by atmospheric positive and negative corona discharges was investigated using a simulation of a wire–cylinder configuration. We provided new insight into the atmospheric corona discharge by introducing a nonlinear force on the charged particles in the vicinity of the wire electrode. To elucidate the origin of both forces in corona discharges, we performed 2D simulations via COMSOL Multiphysics and MATLAB software. It was observed that the direction of nonlinear force is always from the wire to the cylinder regardless of the applied voltage polarity. It was illustrated that the corresponding nonlinear force of the positive corona is larger than that of the negative corona discharge. However, the span of the nonlinear force is greater in the negative corona discharge. The numerical simulation results showed that, in addition to the linear force (Coulomb force), a strong nonlinear force is generated around the wire electrode (powered electrode) that plays a complementary role in the production of electric wind caused by corona discharge. As this nonlinear force is limited to the vicinity of the wire electrode, it is possible to ignore the nonlinear force with a good approximation in the calculation of the total electrohydrodynamic force, but this force cannot be ignored in the process of forming the electric wind.

Degradation of methylene blue using a novel gas-liquid hybrid DDBD reactor: Performance and pathways

A novel gas-liquid hybrid double dielectric barrier discharge (DDBD) reactor with coaxial cylinder configuration was developed for the degradation of methylene blue (MB) in this study. In this DDBD reactor, the reactive species generation occurred in the gas-phase discharge, directly in the liquid, and in the mixture of the working gas bubbles and the liquid, which could effectively increase the contact area between the active substance and MB molecules/intermediates, resulting in an excellent MB degradation efficiency and mineralization (COD and TOC). The electrostatic field simulation analysis by Comsol was carried out to determine the appropriate structural parameters of the DDBD reactor. The effect of discharge voltage, air flow rate, pH, and initial concentration on MB degradation was evaluated. Besides, major oxide species, ·OH, the dissolved O3 and H2O2 generated in this DDBD reactor were determined. Moreover, major MB degradation intermediates were identified by LC-MS, based on which, possible degradation pathways of MB were proposed.

Finite element modeling of partially-confined concrete and RC columns with embedded hexagonal-FRP strips under axial and horizontal loading

Recently, GFRP-strips partial confinement has demonstrated significant revelations as an effective and economical method of structural rehabilitation and strengthening. In this connection, this paper firstly presents a finite element modeling (FEM) on the behavior of a novel confinement configuration of concrete cylinders with embedded hexagonally designed GFRP strips under axial loading using ABAQUS. The simulation outcomes in terms of stress–strain response and damages were experimentally and analytically validated. The second objective of this paper is to develop a FEM to study a new reinforcement method of large scale partially-confined RC columns subjected to a monotonic horizontal loading. For this purpose, full 3D numerical simulations were carried out taking into account the nonlinear behavior of all used materials. The accuracy of the numerical results was emphasized through a comparison with ones of previous experimental tests. In addition, the effects of various parameters on the failure mode and the ultimate lateral carrying load capacity of the columns were studied on twenty-four numerical models. As main conclusion, the predicted results indicated that the proposed strengthening design with a greater encased-GFRP strip thickness combined with smaller spacing significantly improved the performances of RC columns.

Research of dynamic phenomena in a model engine stand

Abstract The highest loads on the elements of the crank–piston system of the internal combustion engine occur during the combustion phase which takes place successively in individual cylinders. Smaller, but also significant, loads occur beyond this phase when dynamic forces begin to play a dominant role. The latter depends on the degree of wear of the engine components. The aim of this article was to analyze the stresses in a model stand for testing dynamic phenomena in an internal combustion engine operating without the combustion process. A stand has been developed which includes a combustion engine driven by an electric motor placed on a mobile frame. The coast-down method was used to estimate the average values of the equivalent moment of inertia of the engine for various cylinder configurations with a lack of compression process due to a significant weakening of the valve springs or significant wear of the valve seats. This lack of compression was simulated on a bench by partially loosening the spark plug mounting. The article presents the values of the determined mean values of the equivalent moment of inertia of the engine and the stress values in the stand frame obtained from its model developed with the use of the finite element method. The obtained values were important from the point of view of the engine dynamics simulation process outside the combustion area and the safe foundation and operation of the stand frame during the experimental tests.

Alfvén continuum in the presence of a magnetic island in a cylinder configuration

In this work, the effect of a magnetic island on Alfvén waves is studied. A physical model is established wherein Alfvén waves propagate in the presence of a magnetic island in a cylindrical geometry. The structure of the Alfvén wave continuum is calculated by considering only the coupling caused by the periodicity in the helical angle of the magnetic island. The results show that the magnetic island can induce an upshift in the Alfvén continuum. Moreover, the coupling between different branches of the continuous spectrum becomes more significant with increasing continuum mode numbers near the boundary of the magnetic island.

Enhancing the Air Conditioning Unit Performance via Energy Storage of Different Inorganic Phase Change Materials with Hybrid Nanoparticles

Air conditioning unit performance, coupled with new configurations of phase change material as thermal energy storage, is investigated in hot climates. During the daytime, the warm exterior air temperature is cooled when flowing over the phase change material structure that was previously solidified by the night ambient air. A theoretical transient model is constructed and solved numerically for the proposed design in plate and cylinder configurations. This model is studied at different inlet hot ambient air temperatures and phase change material types (SP24E and SP26E) without and with inclusion of hybrid nanoparticles. The results affirm that the discharging and charging duration for the cylinder is minimal compared to the plate configuration. Raising the inflow air temperature lowers the exit air temperature and air conditioning coefficient of performance and power-saving but shortens the cooling time. Using phase change material with a relatively low melting temperature increases the melting time and exit air temperature but reduces the charging time. Mixing hybrid nanoparticles with phase change material has a short-term positive influence on air conditioning performance. The maximum power saving for 2 h of working is 16.4% for the cylinder, while for 10 h of working, it is 6.4% for the plate.

Применение унифицированной методологии для аналитического расчёта поглощённых фракций гамма-излучения в биообъектах цилиндрической формы

The paper presents results of application of unified methodology for analytical evaluation of absorbed radiation fractions for dosimetry of cylinder-shape bio-objects following the internal uniform contamination with emitted photons. New models have been used for calculation of photon fractions dose absorbed by various non-human biological objects of various cylinder configurations and sizes. Evaluated absorbed photon radiation doses are in accordance with independently tested data, obtained by numeric integration of basic absorbed photon radiation dose with account of the cylinder volume. The theoretical interpretation of used calculation formulae based on the theory of middle chords in convex bodies is given in the paper. An advantage of the unified method is the possibility to use simple algebraic formulae for calculating absorbed radiation dose fractions in convex bodies (spheres, ellipsoids, cylinders) without the use of Monte Carlo computer programs and adjustment parameters. In contrast to the known European computer complex ERICA Tool the developed method of calculation of radiation doses to cylinder shape bio-objects may be also used for dosimetry of non-human species, for express-assessment of radioecological situation in radionuclide-contaminated areas, as well as for radiation protection of biota.

Numerical modeling of a wire mesh for aerodynamic noise reduction

A novel wire mesh consisting of very fine wires and pores is numerically investigated for the purpose of noise reduction. To develop a numerical model for this wire mesh, a set of experimental flow-field data has been deployed for the model validation. The experimental data were measured with only 22% of the wind-tunnel cross section covered by the wire mesh, taking into account the vortex shedding from both sides of the wire-mesh fairing. It is found that existing wire-mesh models using a damping-type source term proportional to the square of flow velocity do not perform well in modeling this novel wire mesh. To tackle this issue, an improvement is proposed by additionally introducing a linear term to account for the permeability of the wire mesh, based on another set of experiments with the wind-tunnel cross section fully covered by the wire mesh. The proposed model is then validated against the experimental data, demonstrating its capability in modeling the wire mesh. Subsequently, the model is applied to a tandem cylinder configuration. Results show that a wide but short-span wire mesh significantly reduces the dominant tone of tandem cylinders, noise at higher frequencies, as well as the overall sound pressure levels.

Regulation of the fabric wet adhesion property through structural design of desirable scale: A theoretical suggestion

It is a universal phenomenon existing in both manufacture and everyday life that fabrics tend to adhere under wet condition. The phenomenon is closely related to the interaction of the fabric surface structure and the liquid. Among the complex structures and numerous varieties of fabrics, the features of typical fabric structures are employed to establish a geometrical model (including structural configuration of spherical protrusion, cylinder, loop, and pore) in order to investigate wet adhesion property under the existence of massive liquid media. Through mechanical analysis on material-liquid bridge system (consisting of fabric material, liquid bridge adhering to the material, and the solid-liquid interface), a quantitative computational theoretical model is built with parameters including critical distance [Formula: see text], radius of the structure R*, apparent receding contact angle [Formula: see text], wetted sphere angle [Formula: see text], and so on, finding the rules and proposing a series of preferred ranges of regulating structural parameters to effectively achieve desirable fabric wet adhesion property in different circumstances. The research findings could be applied to the scientific prediction and on demand regulation of fabric wet adhesion property. They are of great significance with regard to improving the convenience of fabric wet processing, application properties under wet environments, apparel health and comfort for human body, and so on.

Effect of inner cylinder configuration on liquid droplet dispersion and size distributions in a Taylor-Couette reactor

Effect of varying surface structures of the inner rotating cylinder on characteristics of droplet formed in a liquid–liquid Taylor-Couette (TC) reactor is investigated. Two novel surface structures of inner cylinders, designed with axially corrugated surface (N40) and with typical three-dimensional roughness (NZ40), respectively, were adopted for the investigation. The high-speed camera visualization was used to measure the droplet size while CFD modelling was applied to predict the fluid dynamics of TC flows with such two inner cylinders and a smooth surface inner cylinder, thus revealing the effect of inner cylinder configuration on the droplet formation in the TC flows. The experimental results have clearly indicated that both the rotation speed and surface configuration variation of the inner cylinders affect the generated droplet size and distribution. As the rotation speed increases, the droplet size is obviously reduced for all types of inner cylinder configurations. Compared with the conventional smooth surface inner cylinder, the two new types of inner cylinders with special surface structures can produce much smaller droplets, especially for the case of N40 surface structure inner cylinder, with which the smallest droplet size was found at all the experimental conditions. CFD modelling results have demonstrated that the surface configurations of inner cylinder can significantly affect the turbulence generation, consequently altering the distributions of turbulent kinetic energy and local turbulence shear strain rate. As a result, the droplet size was found to be controlled by the change of the local turbulent dissipation rate that was strongly associated with the surface structures of inner cylinders. Empirical correlations that correlate the droplet size with the Weber number and Reynolds number were proposed based on the high-speed camera visualization data.

Thermal Performance in Convection Flow of Nanofluids Using a Deep Convolutional Neural Network

This study develops a geometry adaptive, physical field predictor for the combined forced and natural convection flow of a nanofluid in horizontal single or double-inner cylinder annular pipes with various inner cylinder sizes and placements based on deep learning. The predictor is built with a convolutional-deconvolutional structure, where the input is the annulus cross-section geometry and the output is the temperature and the Nusselt number for the nanofluid-filled annulus. Profiting from the proven ability of dealing with pixel-like data, the convolutional neural network (CNN)-based predictor enables an accurate end-to-end mapping from the geometry input and the desired nanofluid physical field. Taking the computational fluid dynamics (CFD) calculation as the basis of our approach, the obtained results show that the average accuracy of the predicted temperature field and the coefficient of determination R2 are more than 99.9% and 0.998 accurate for single-inner cylinder nanofluid-filled annulus; while for the more complex case of double-inner cylinder, the results are still very close, higher than 99.8% and 0.99, respectively. Furthermore, the predictor takes only 0.038 s for each nanofluid field prediction, four orders of magnitude faster than the numerical simulation. The high accuracy and the fast speed estimation of the proposed predictor show the great potential of this approach to perform efficient inner cylinder configuration design and optimization for nanofluid-filled annulus.

PO23 Presentation Time: 7:40 AM: Dosimetric Effects of Model-Based TG-186 Vaginal Cylinder Treatment Plans

<h3>Purpose</h3> The vaginal cuff is the main location of relapse after curative surgery for early-stage endometrial cancer. Adjuvant vaginal cylinder brachytherapy is increasingly used to decrease the risk of postoperative recurrence due to low vaginal cuff recurrence rates and low gastrointestinal toxicity. Current clinical treatment planning follows TG-43 guidance, which assumes that all material is water. Cylinder composition can vary substantially from water equivalent materials, highlighting the importance of investigations into the dosimetric impact of cylinder composition. In the present study, we evaluated dosimetric characteristics of plans utilizing TG-186, which provides a model-based calculation approach to account for heterogeneities in cylinder configuration. <h3>Materials and Methods</h3> Ten graphically planned vaginal cylinder plans were retrospectively evaluated in this study. Each patient was fitted with an Elekta cylinder (part# 084350) comprised primarily of polysulfone (PSU, 1.29g/cc) and secondarily of stainless steel (8.00g/cc) for the stem and rings. Cylinder diameters ranged from 2.5cm to 3.5cm, and treatment lengths from 4.0cm to 7.5cm. Dose prescription was 7Gy/fx x 3fx at 5mm depth for the most proximal 2cm of the cylinder and at cylinder surface for the remainder of the treatment length. Treatment planning was conducted in Oncentra Brachy 4.5.3. The original TG-43 plans (which assume a water-equivalent material, 1.00g/cc) were recalculated using Oncentra's Advanced Collapsed-Cone Engine (ACE) according to TG-186's guidance to account for differences in density. Dwell times were unchanged, so heterogeneity correction accounts for any dosimetric changes. Points placed at the middle of the 5mm depth portion and the middle of the surface depth portion - prescription dose for TG-43 - were compared, as were rectum points. In addition, all isodose lines were converted to contours and compared for the following metrics: percent volume change, Hausdorff distance (HD), mean distance to agreement (MDA), and Dice coefficient. A paired Wilcoxon signed-rank test assessed for significance. <h3>Results</h3> Recalculated TG-186 plans were found to have consistently and statistically significantly (p<0.05) less dose: 5mm depth: -10.5% +/- 2.4%, surface: -15.9% +/- 3.5%, and rectum: -12.1% +/- 3.4% average dose reduction. Compared to TG-43, the TG-186 100% isodose lines as contours show an average volume decrease of 22.2% +/- 2.5%, HD 4.6mm +/- 1.2mm, MDA 1.5mm +/- 0.3mm, and Dice 0.87 +/- 0.02. The high dose region as defined by 150% of prescription (25.7% volume decrease, HD 4.4mm, MDA 1.3mm) and the low dose region of 50% prescription (19.5% volume decrease, HD 8.2mm, MDA 1.9mm) show similar trends. <h3>Conclusions</h3> Ten TG-43 graphical vaginal cylinder plans were recalculated using TG-186 to account for the material composition of the cylinders. ACE calculations show that the dose delivered to target prescription points and to OARs may be less than originally planned, although the impact on treatment efficacy is presently unclear. Further studies on other cylinder models will be useful to evaluate the generalizability of these findings, as well as any potential effect on treatment.

Estimation of Hydrodynamic Forces on Cylinders Undergoing Flow-Induced Vibrations Based on Modal Analysis

Abstract The objective of the present study is estimating hydrodynamic forces acting on cylinders undergoing vortex-induced vibration (VIV) using dynamic mode decomposition (DMD). The cylinders are subjected to a uniform incoming flow at a laminar Reynolds number (Re = 250) and an upper transition Reynolds number (Re = 3.6 × 106) (Re = U∞D/ν defined based on the incoming flow U∞, the diameter of the cylinder D, and the viscosity of the fluid ν). Both a single cylinder and a configuration of piggyback cylinders are considered. Numerical simulations based on two-dimensional unsteady Reynolds-averaged Navier–Stokes (URANS) equations combined with the k−ω SST turbulence model are carried out to obtain the snapshots of the surrounding flow fields for DMD analysis. The DMD method is a powerful tool to obtain the spatial–temporal evolution characteristics of the coherent structures in the wake flow behind the cylinders. In the present study, this modal decomposition method is combined with a moving reference frame around the cylinders. The dominant DMD modes with their corresponding frequencies of the wake flows are identified and are used to reconstruct the flow fields. The large-scale shedding vortices are captured by the dominant modes. The reconstructed wake flow behind the cylinders is used to estimate the drag and lift forces on the cylinders combined with a force partitioning analysis.

Empirical Model of Uniflow Scavenging for Ultra-Long-Two-Stroke Marine Diesel Engines

_ Low-speed two-stroke diesel engines are widely employed in the marine industry, especially in tanks, owing to their advantages of fuel economy and reliability. However, with the emerging ultra-long-stroke trend, existing scavenging models, which are based on the configuration of cylinders with short strokes, are no longer applicable. In this study, we investigate the flow field and residual exhaust gas distribution in a cylinder using particle image velocimetry and computational fluid dynamics (CFD) simulations. The result shows a strong Burgers vortex structure upstream of the scavenger flow and dissipates gradually as it moves downstream. Furthermore, the scavenging process comprises three processes according to the detailed CFD analysis: displacement, mixing scavenging, and short circuit. Inspired by the results, a tailored empirical model of ultra-long-stroke uniflow scavenging comprising three sub-models is proposed. Specifically, a logarithmic relationship between the concentration level and scavenging deliver ratio is proposed to describe the mixing scavenging process. Finally, the model was validated against CFD results. The results demonstrate that the discrepancy in the scavenging efficiency curve predicted by this model and CFD is less than 1%, thereby demonstrating model reliability. Introduction Owing to the advantages of a large power range, high thermal efficiency, low fuel consumption rate, and good reliability, marine low-speed engines are widely used to provide power to civil ships (Heywood 1999). The market for low-speed engines is vast and is improving the performance of low-speed engines through research has great economic and environmental significance (Woodyard 2004). To meet the requirements of ship owners for lower fuel consumption and the IMO’s regulation of halving the greenhouse gas emissions of newly built ships, the ultra-long stroke of low-speed engines has become a trend (Lamas &amp; Vidal 2012). With an ultralong stroke, the combustion speed of low-quality fuel oil is slow, and an ultra-long-stroke cylinder can prolong the expansion process, improve the combustion process, and reduce the fuel consumption rate (Fenghua 2014). The ultra-long-stroke diesel engine further creates fuel savings of 3.5–7% based on the original low fuel consumption. Previously, ultra-long strokes could not be achieved, limited by materials and manufacturing processes. However, in recent years, with the advancement of materials and processes, ultra-long strokes have been widely adopted as they have demonstrated superior competitiveness.

Numerical methodology to evaluate unipolar saturation current limit of DC corona discharge in complex geometries

The unipolar saturation current limit ({I}_{sat}) gives an upper limit to the corona current that can be obtained from a unipolar corona discharge. Therefore, it implies a theoretical limit to the performance of unipolar corona discharge devices. However, it has not been widely used in practice because it is difficult to deal with complex discharge configurations in an analytical way. This study aims to establish and validate a numerical methodology to evaluate the maximum current, which numerically imitates the unipolar saturation current limit. It was shown that the maximum current has the same mathematical definition as the unipolar saturation current. For validation, the maximum current was compared with an analytical solution of the Poisson equation for the coaxial cylinders configuration. The differences between the maximum current and unipolar saturation current limit for the coaxial cylinders, pin-to-plane, and single wire-to-plane configurations were discussed in terms of the assumptions used in the semi-analytical derivation of the unipolar saturation current limit. The validated methodology was applied to a multiple wire-to-plane configuration, for which a semi-analytical expression of the unipolar saturation current limit has not yet been developed. The effects of geometric and operation parameters on the maximum currents of the multiple wire-to-plane configuration were analyzed. The results were regressed into a single formula.

Experimentation on Optimal Configuration and Size of Thin Cylinders in Natural Convection

In this paper, an experimental study of laminar, steady state natural convection heat transfer from heated thin cylinders in an infinite air medium has been reported. Two electrically heated cylinders having the same slenderness ratio (L/D) i.e. 6.1 but different diameters i.e. 3.8 cm and 5.08 cm were used. 105 experiments were carried out to study the effect of diameter and inclination angle of thin cylinder on natural convection heat transfer. After mandatory corrections of radiation and endcap heat losses, convective heat transfer results were presented in the form of local and average dimensionless numbers. For vertical configuration of thin cylinder, Nusselt number was varied from 52.99 to 95.10 corresponding to 1.28×108≤Ra*L≤1.08×1010. While for horizontal configuration,Nusselt number was varied from 10.74 to 17.78 corresponding to 9.42×104≤Ra*D≤8.17×106. Results were compared with the published data and found satisfactory as the maximum percentage difference was only 3.09%. The essence of research is that the heat transfer coefficient increases with decrease in diameter and increase in inclination angle. Smoke flow visualization was done to capture patterns of fluid flow. Finally, comparison was made to quantify increase in Nusselt number from slender cylinder as compared to the flat plate.

Numerical Study of the Effect of a Heated Cylinder on Natural Convection in a Square Cavity in the Presence of a Magnetic Field

The present research was developed to find out the effect of heated cylinder configurations in accordance with the magnetic field on the natural convective flow within a square cavity. In the cavity, four types of configurations—left bottom heated cylinder (LBC), right bottom heated cylinder (RBC), left top heated cylinder (LTC) and right top heated cylinder (RTC)—were considered in the investigation. The current mathematical problem was formulated using the non-linear governing equations and then solved by engaging the process of Galerkin weighted residuals based on the finite element scheme (FES). The investigation of the present problem was conducted using numerous parameters: the Rayleigh number (Ra = 103–105), the Hartmann number (Ha = 0–200) at Pr = 0.71 on the flow field, thermal pattern and the variation of heat inside the enclosure. The clarifications of the numerical result were exhibited in the form of streamlines, isotherms, velocity profiles and temperature profiles, local and mean Nusselt number, along with heated cylinder configurations. From the obtained outcomes, it was observed that the rate of heat transport, as well as the local Nusselt number, decreased for the LBC and LTC configurations, but increased for the RBC and RTC configurations with the increase of the Hartmann number within the square cavity. In addition, the mean Nusselt number for the LBC, RBC, LTC and RTC configurations increased when the Hartmann number was absent, but decreased when the Hartmann number increased in the cavity. The computational results were verified in relation to a published work and were found to be in good agreement.

Vortex Dislocation In The Near Wake Of A Cylinder With Span-Wise Variations In Diameter

We examined the evolution of three-dimensional vortex shedding patterns induced by spanwise variations of the cylinder diameter. Two distinct types of shedding patterns have identified through flow visualization: continuous (in-phase) oblique shedding where vortices shed with lower frequency stay attached to the vortices with higher frequency without any discontinuity or splitting and discontinuous (out-of-phase) shedding where the lower frequency vortices have no attachment to higher frequency vortices and vortex dislocation occurs. The dislocation seen in the flow is strongly influenced by the span wise irregularities. We observed a clear and strong in-phase spanwise vortex shedding for the three smooth cylinder configurations tested in the study. The tapered, bumps and steps configurations showed sections of strong coherent spanwise vortex shedding, and we identified hardly any coherent structures for the rib, sinusoidal, and helical configurations. Depending on the geometry and number of span wise irregularities, incoherent structures make difficult to determine the occurrence and location of vortex dislocations in the cylinder wake without a method that enables a reduction in the complexity. Here, we introduce a numerical approach that obtains the dominant structures of vortex shedding patterns by reducing the complexity in noisy data. The method maps the variations in the oblique shedding angle over time and provide quantitative conclusions on the intermittent occurrence and location of vortex dislocations in the 15000 snapshots taken for each of the different cylinder geometries regardless of turbulence.

Acoustic metamaterial design using Conditional Wasserstein Generative Adversarial Networks

This talk presents a method for generating planar configurations of scatterers with a reduced total scattering cross section (TSCS) by means of generative modeling and deep learning. The TSCS minimization via repeated forward modeling techniques, trial-error methods, and traditional optimization methods requires considerable computer resources and time. However, similar or better results can be achieved more efficiently by training a deep learning model to generate such optimized configurations producing low scattering effect. In this work, the Conditional Wasserstein Generative Adversarial Networks (cWGAN) is combined with Convolutional Neural Networks (CNN) to create the generative modeling architecture [1]. The generative model is implemented with a conditional proponent to allow the TSCS targeted design generation and is enhanced with the coordinate convolution (CordConv) layer to improve the model’s spatial recognition of cylinder configurations. The cWGAN model [1] is capable of generating images of 2D configurations of scatterers that exhibit low scattering. The method is demonstrated by giving examples of generating 2-cylinder and 4-cylinder planar configurations with minimal TSCS. [1]. P. Lai, F. Amirkulova, and P. Gerstoft. “Conditional Wasserstein generative adversarial networks applied to acoustic metamaterial design,” J. Acoust. Soc. Am. 150(6), 4362–4374 (2021).